You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@netbeans.apache.org by Geertjan Wielenga <ge...@googlemail.com> on 2017/10/04 12:31:35 UTC
New pattern for license converter tool
Hi all,
Yesterday at JavaOne, some of us got together to work on the Modules Review
and found a new pattern that needs to be handled in the converter tool.
The pattern can be found in several hundred files, so we can clean these
all up at the same time by twesking the tool for this.
Here are three places where this pattern is found:
- api.progress.nb/apichanges.xml
- api.progress/apichanges.xml
- api.progress.nb/apichanges.xml
Except, we haven't been able to figure out yet how the pattern (it is a
variation on the standard Oracle license) is different and why it hasn't
been handled by the tool yet.
Could Jan or someone interested in this question else take a look?
Geertjan
Re: New pattern for license converter tool
Posted by Jan Lahoda <la...@gmail.com>.
I've just opened:
https://github.com/apache/incubator-netbeans/pull/72
(files from active pull requests have been excluded from the patch; the
patch rewrites 48 files.)
Jan
On Thu, Oct 5, 2017 at 3:43 PM, Geertjan Wielenga <
geertjan.wielenga@googlemail.com> wrote:
> Can you/have you run the tool again, with these changes, and pushed the
> changed license headers?
>
> And must all the other pull requests be merged first because in some cases
> people might have manually changed some of these headers and there could be
> problems in those cases?
>
> Thanks,
>
> Gj
>
> On Wed, Oct 4, 2017 at 2:29 PM, Jan Lahoda <la...@gmail.com> wrote:
>
> > FWIW, I've changed the tool to accept license headers in the second
> comment
> > in the file, and changed the limit where the comment must start to 300.
> > This seems to cover about 63 additional files.
> >
> > For the examples-readme.txt, the comment/header detection in such files
> is
> > tricky. I think the reason is that the second line in the file does not
> > start with the comment character ('#'), so the detection fails.
> >
> > Jan
> >
> > On Wed, Oct 4, 2017 at 2:49 PM, Geertjan Wielenga <
> > geertjan.wielenga@googlemail.com> wrote:
> >
> > > Great!
> > >
> > > Another similar one is in api.visual/examples-readme.txt.
> > >
> > > Gj
> > >
> > > On Wed, Oct 4, 2017 at 5:36 AM, Jan Lahoda <la...@gmail.com> wrote:
> > >
> > > > I think there are two problem with these files:
> > > > -the tool only looks for the first comment, and the headers in these
> > > files
> > > > are in the second comment, so the tool won't find it.
> > > > -the tool only looks for comments that begin in the first 100
> > characters
> > > > from the beginning (a safety feature, as a few files contain texts of
> > the
> > > > license headers as patterns)
> > > >
> > > > I'll look at those.
> > > >
> > > > Jan
> > > >
> > > >
> > > > On Wed, Oct 4, 2017 at 2:31 PM, Geertjan Wielenga <
> > > > geertjan.wielenga@googlemail.com> wrote:
> > > >
> > > > > Hi all,
> > > > >
> > > > > Yesterday at JavaOne, some of us got together to work on the
> Modules
> > > > Review
> > > > > and found a new pattern that needs to be handled in the converter
> > tool.
> > > > >
> > > > > The pattern can be found in several hundred files, so we can clean
> > > these
> > > > > all up at the same time by twesking the tool for this.
> > > > >
> > > > > Here are three places where this pattern is found:
> > > > >
> > > > >
> > > > > - api.progress.nb/apichanges.xml
> > > > > - api.progress/apichanges.xml
> > > > > - api.progress.nb/apichanges.xml
> > > > >
> > > > >
> > > > > Except, we haven't been able to figure out yet how the pattern (it
> > is a
> > > > > variation on the standard Oracle license) is different and why it
> > > hasn't
> > > > > been handled by the tool yet.
> > > > >
> > > > > Could Jan or someone interested in this question else take a look?
> > > > >
> > > > > Geertjan
> > > > >
> > > >
> > >
> >
>
Re: New pattern for license converter tool
Posted by Geertjan Wielenga <ge...@googlemail.com>.
Can you/have you run the tool again, with these changes, and pushed the
changed license headers?
And must all the other pull requests be merged first because in some cases
people might have manually changed some of these headers and there could be
problems in those cases?
Thanks,
Gj
On Wed, Oct 4, 2017 at 2:29 PM, Jan Lahoda <la...@gmail.com> wrote:
> FWIW, I've changed the tool to accept license headers in the second comment
> in the file, and changed the limit where the comment must start to 300.
> This seems to cover about 63 additional files.
>
> For the examples-readme.txt, the comment/header detection in such files is
> tricky. I think the reason is that the second line in the file does not
> start with the comment character ('#'), so the detection fails.
>
> Jan
>
> On Wed, Oct 4, 2017 at 2:49 PM, Geertjan Wielenga <
> geertjan.wielenga@googlemail.com> wrote:
>
> > Great!
> >
> > Another similar one is in api.visual/examples-readme.txt.
> >
> > Gj
> >
> > On Wed, Oct 4, 2017 at 5:36 AM, Jan Lahoda <la...@gmail.com> wrote:
> >
> > > I think there are two problem with these files:
> > > -the tool only looks for the first comment, and the headers in these
> > files
> > > are in the second comment, so the tool won't find it.
> > > -the tool only looks for comments that begin in the first 100
> characters
> > > from the beginning (a safety feature, as a few files contain texts of
> the
> > > license headers as patterns)
> > >
> > > I'll look at those.
> > >
> > > Jan
> > >
> > >
> > > On Wed, Oct 4, 2017 at 2:31 PM, Geertjan Wielenga <
> > > geertjan.wielenga@googlemail.com> wrote:
> > >
> > > > Hi all,
> > > >
> > > > Yesterday at JavaOne, some of us got together to work on the Modules
> > > Review
> > > > and found a new pattern that needs to be handled in the converter
> tool.
> > > >
> > > > The pattern can be found in several hundred files, so we can clean
> > these
> > > > all up at the same time by twesking the tool for this.
> > > >
> > > > Here are three places where this pattern is found:
> > > >
> > > >
> > > > - api.progress.nb/apichanges.xml
> > > > - api.progress/apichanges.xml
> > > > - api.progress.nb/apichanges.xml
> > > >
> > > >
> > > > Except, we haven't been able to figure out yet how the pattern (it
> is a
> > > > variation on the standard Oracle license) is different and why it
> > hasn't
> > > > been handled by the tool yet.
> > > >
> > > > Could Jan or someone interested in this question else take a look?
> > > >
> > > > Geertjan
> > > >
> > >
> >
>
Re: New pattern for license converter tool
Posted by Jan Lahoda <la...@gmail.com>.
FWIW, I've changed the tool to accept license headers in the second comment
in the file, and changed the limit where the comment must start to 300.
This seems to cover about 63 additional files.
For the examples-readme.txt, the comment/header detection in such files is
tricky. I think the reason is that the second line in the file does not
start with the comment character ('#'), so the detection fails.
Jan
On Wed, Oct 4, 2017 at 2:49 PM, Geertjan Wielenga <
geertjan.wielenga@googlemail.com> wrote:
> Great!
>
> Another similar one is in api.visual/examples-readme.txt.
>
> Gj
>
> On Wed, Oct 4, 2017 at 5:36 AM, Jan Lahoda <la...@gmail.com> wrote:
>
> > I think there are two problem with these files:
> > -the tool only looks for the first comment, and the headers in these
> files
> > are in the second comment, so the tool won't find it.
> > -the tool only looks for comments that begin in the first 100 characters
> > from the beginning (a safety feature, as a few files contain texts of the
> > license headers as patterns)
> >
> > I'll look at those.
> >
> > Jan
> >
> >
> > On Wed, Oct 4, 2017 at 2:31 PM, Geertjan Wielenga <
> > geertjan.wielenga@googlemail.com> wrote:
> >
> > > Hi all,
> > >
> > > Yesterday at JavaOne, some of us got together to work on the Modules
> > Review
> > > and found a new pattern that needs to be handled in the converter tool.
> > >
> > > The pattern can be found in several hundred files, so we can clean
> these
> > > all up at the same time by twesking the tool for this.
> > >
> > > Here are three places where this pattern is found:
> > >
> > >
> > > - api.progress.nb/apichanges.xml
> > > - api.progress/apichanges.xml
> > > - api.progress.nb/apichanges.xml
> > >
> > >
> > > Except, we haven't been able to figure out yet how the pattern (it is a
> > > variation on the standard Oracle license) is different and why it
> hasn't
> > > been handled by the tool yet.
> > >
> > > Could Jan or someone interested in this question else take a look?
> > >
> > > Geertjan
> > >
> >
>
Re: New pattern for license converter tool
Posted by Geertjan Wielenga <ge...@googlemail.com>.
Great!
Another similar one is in api.visual/examples-readme.txt.
Gj
On Wed, Oct 4, 2017 at 5:36 AM, Jan Lahoda <la...@gmail.com> wrote:
> I think there are two problem with these files:
> -the tool only looks for the first comment, and the headers in these files
> are in the second comment, so the tool won't find it.
> -the tool only looks for comments that begin in the first 100 characters
> from the beginning (a safety feature, as a few files contain texts of the
> license headers as patterns)
>
> I'll look at those.
>
> Jan
>
>
> On Wed, Oct 4, 2017 at 2:31 PM, Geertjan Wielenga <
> geertjan.wielenga@googlemail.com> wrote:
>
> > Hi all,
> >
> > Yesterday at JavaOne, some of us got together to work on the Modules
> Review
> > and found a new pattern that needs to be handled in the converter tool.
> >
> > The pattern can be found in several hundred files, so we can clean these
> > all up at the same time by twesking the tool for this.
> >
> > Here are three places where this pattern is found:
> >
> >
> > - api.progress.nb/apichanges.xml
> > - api.progress/apichanges.xml
> > - api.progress.nb/apichanges.xml
> >
> >
> > Except, we haven't been able to figure out yet how the pattern (it is a
> > variation on the standard Oracle license) is different and why it hasn't
> > been handled by the tool yet.
> >
> > Could Jan or someone interested in this question else take a look?
> >
> > Geertjan
> >
>
Re: New pattern for license converter tool
Posted by Jan Lahoda <la...@gmail.com>.
I think there are two problem with these files:
-the tool only looks for the first comment, and the headers in these files
are in the second comment, so the tool won't find it.
-the tool only looks for comments that begin in the first 100 characters
from the beginning (a safety feature, as a few files contain texts of the
license headers as patterns)
I'll look at those.
Jan
On Wed, Oct 4, 2017 at 2:31 PM, Geertjan Wielenga <
geertjan.wielenga@googlemail.com> wrote:
> Hi all,
>
> Yesterday at JavaOne, some of us got together to work on the Modules Review
> and found a new pattern that needs to be handled in the converter tool.
>
> The pattern can be found in several hundred files, so we can clean these
> all up at the same time by twesking the tool for this.
>
> Here are three places where this pattern is found:
>
>
> - api.progress.nb/apichanges.xml
> - api.progress/apichanges.xml
> - api.progress.nb/apichanges.xml
>
>
> Except, we haven't been able to figure out yet how the pattern (it is a
> variation on the standard Oracle license) is different and why it hasn't
> been handled by the tool yet.
>
> Could Jan or someone interested in this question else take a look?
>
> Geertjan
>